Try to calculate the single-point energy of the two hydrogen-bonded species separately, then together linked with hydrogen bond(s) apllying BSSE correction.

From these calculations you'll get energy of the hydrogen bond(s) with the following formula: Ehbond = Ea +Eb - Eab(BSSE corrected) 

The energy of the UV photon what you use is 78.3 kcal/mol at 365 nm so if the calculated hydrogen bond(s) is weaker than this, it may cleave during irradiation, however it heavily depends on the molecular structure, but I hope this will help.

Good answer above. To add to it: each reaction can in theory go forward and backwards. If the energy (actually Gibbs free energy) of reactants and products (in the example above the energy of cis and trans isomer) are similar, the process will be most likely reversible, as the energy barriers for reaction going forward and backwards will be similar. There will be 

If the GIbbs free energies of reactants and products will be relatively far from each other, the reaction will be most likely irreversible and will proceed to the species with the lower Gibbs free energy.

Hello Krisztián Csankó and Bartosz Trzaskowski,

Thank you for your replies. I agree with Krisztián's answer. It would have been much easier if it was a dimer in solution. Since I have to simulate this effect in crystalline sample, it will greatly depend upon not only the molecular structure but also the symmetry of the crystal. I will see if the same calculations can be done using periodic boundary conditions.

As for the reversibility of the reaction, I would like to ask if the free energy based calculations hold true for reactions in crystalline state as they do for the reactions in solution state. Because here, the same reaction is assisted or hindered by other molecules arranged in crystalline symmetry. Could you please elaborate upon or give a reference for reactions in crystalline state?